![\"A](\"https:\/\/www.newsbeep.com\/il\/wp-content\/uploads\/2026\/04\/shijie_lyu_300x.jpg\")
\nShijie Lu<\/p>\n
\nShijie Lu<\/p>\n
Even with all the hype around streaming, traditional TV still dominates ad spend. Advertisers are putting $139 billion into linear ads this year, compared to just $33 billion for ads on streaming\/connected TV.<\/p>\n
With no way to track individual behavior among traditional TV viewers, it\u2019s difficult to determine whether all that spending gets results. New research from the University of Notre Dame helps determine the return on investment for TV ads, ironically by using digital data. By combining massive datasets that track exactly what households watch and buy second by second, the study separates the real impact of TV ads from other factors.<\/p>\n
Traditional methods of measurement, which rely mostly on ratings and aggregate market data, appear to overestimate ad effectiveness by 55 percent in a study of advertising for food delivery services, according to Shijie Lu<\/a>, the Howard J. and Geraldine F. Korth Associate Professor of Marketing at Notre Dame\u2019s Mendoza College of Business<\/a>. Lu\u2019s research, \u201cLeveraging Large-Scale Granular Single-Source Data for TV Advertising: An Identification Strategy<\/a>,\u201d is published online in Marketing Science.<\/p>\n Imagine that a household watches only part of a live game. If a food delivery ad airs during the portion they watched, they may see it; if it airs earlier or later, they may miss it. That timing difference creates a kind of natural experiment, helping the researchers isolate the ad\u2019s true effect from other factors, such as which households were already more likely to order food. Researchers could not easily do this before with traditional TV measurement. Smart TV tracking now provides second-by-second household viewing data, making this kind of measurement possible at a much finer level.<\/p>\n Using LG smart TV data, Lu and co-authors Tsung-Yiou Hsieh from Oklahoma State University and Rex Yuxing Du from the University of Texas at Austin analyzed the viewing habits of millions of people who opted in to sharing their viewing data, letting the researchers see exactly what was on peoples\u2019 screens \u2014 broadcast networks such as NBC and ABC, specifically \u2014 over a four-month period. The study didn\u2019t track streaming apps like Hulu or Amazon. LG watched what viewers watched and connected that data to people\u2019s food delivery app usage to measure ad impact.<\/p>\n \u201cThis is a game-changer,\u201d Lu said, \u201cbecause we can now link precise TV viewing data with real purchase history to measure TV ad effectiveness more credibly.<\/p>\n \u201cBrands are overestimating their campaigns and wasting money on ineffective placements,\u201d he said. \u201cWe show TV ads are only about half as effective as we thought. When corrected, the real sales impact is much lower, which has important implications for how advertisers evaluate performance and allocate spending.\u201d<\/p>\n In addition to showing that traditional measures greatly overstated the effects of TV ads, the new measurement method revealed additional insights that could help companies better target their ads.<\/p>\n Data show that promotions for first-time buyers increase retention. Viewers\u2019 responsiveness to ads peaks within two days of purchasing food on a delivery app, with the highest engagement rate found among customers who have ordered two to four times previously. Young, tech-savvy sports fans are better prospects than older news viewers.<\/p>\n \u201cThe old ways of measuring TV ads are missing an important part of the picture, because they do not fully account for who is more likely to see ads and who is more likely to buy,\u201d Lu said.<\/p>\n Traditional TV ad tracking confuses ad effectiveness with pre-existing habits (like who is already likely to buy or who watches a lot of TV), leading to inflated results. This research fixes that by isolating the random timing of ad slots within shows, allowing the team to accurately measure the true sales lift of TV ads and determine how that impact varies based on a customer\u2019s history.<\/p>\n The study provides a powerful tool for more precisely measuring the return on investment of TV advertising. By targeting ads based on what viewers actually buy \u2014 not just demographics like age or gender \u2014 this approach brings digital-level precision to TV.<\/p>\n